Update app.py

app.py

@@ -38,7 +38,7 @@ class MyModel:
             # )
             # FastLanguageModel.for_inference(client) # Enable native 2x faster inference
             tokenizer = AutoTokenizer.from_pretrained(model)
-            client = AutoPeftModelForCausalLM.from_pretrained(model, load_in_4bit=True)
+            client = AutoPeftModelForCausalLM.from_pretrained(model) #, load_in_4bit=True)
 
             self.client = client
             self.tokenizer = tokenizer